Molecular mechanisms of bone formation in spondyloarthritis

Abstract

Spondyloarthritis comprise a group of inflammatory rheumatic diseases characterized by its association to HLA-B27 and the presence of arthritis and enthesitis. The pathogenesis involves both an inflammatory process and new bone formation, which eventually lead to ankylosis of the spine. To date, the intrinsic mechanisms of the pathogenic process have not been fully elucidated, and our progress is remarkable in the identification of therapeutic targets to achieve the control of the inflammatory process, yet our ability to inhibit the excessive bone formation is still insufficient. The study of new bone formation in spondyloarthritis has been mostly conducted in animal models of the disease and only few experiments have been done using human biopsies. The deregulation and overexpression of molecules involved in the osteogenesis process have been observed in bone cells, mesenchymal cells, and fibroblasts. The signaling associated to the excessive bone formation is congruent with those involved in the physiological processes of bone remodeling. Bone morphogenetic proteins and Wnt pathways have been found deregulated in this disease; however, the cause for uncontrolled stimulation remains unknown. Mechanical stress appears to play an important role in the pathological osteogenesis process; nevertheless, the association of other important factors, such as the presence of HLA-B27 and environmental factors, remains uncertain. The present review summarizes the experimental findings that describe the signaling pathways involved in the new bone formation process in spondyloarthritis in animal models and in human biopsies. The role of mechanical stress as the trigger of these pathways is also reviewed.

Full text

Please
cite
this
article
in
press
as:
González-Chávez
SA,
et
al.
Molecular
mechanisms
of
bone
formation
in
spondyloarthritis.
Joint
Bone
Spine
(2016),
http://dx.doi.org/10.1016/j.jbspin.2015.07.008
ARTICLE IN PRESS
G Model
BONSOI-4290;
No.
of
Pages
7
Joint
Bone
Spine
xxx
(2016)
xxx–xxx
Available
online
at
ScienceDirect
www.sciencedirect.com
Review
Molecular
mechanisms
of
bone
formation
in
spondyloarthritis
Susana
Aideé
González-Cháveza,b,
Celia
María
Qui˜
nonez-Floresa,b,
César
Pacheco-Tenaa,∗
aFacultad
de
Medicina,
Universidad
Autónoma
de
Chihuahua,
Circuito
Universitario
Campus
II,
31125
Chihuahua,
CP,
Mexico
bFacultad
de
Ciencias
de
la
Cultura
Física,
Universidad
Autónoma
de
Chihuahua,
31125
Chihuahua,
CP,
Mexico
a
r
t
i
c
l
e
i
n
f
o
Article
history:
Accepted
20
July
2015
Available
online
xxx
Keywords:
Ankylosis
Osteogenesis
Mechanosensing
Signaling
pathways
a
b
s
t
r
a
c
t
Spondyloarthritis
comprise
a
group
of
inflammatory
rheumatic
diseases
characterized
by
its
association
to
HLA-B27
and
the
presence
of
arthritis
and
enthesitis.
The
pathogenesis
involves
both
an
inflammatory
process
and
new
bone
formation,
which
eventually
lead
to
ankylosis
of
the
spine.
To
date,
the
intrinsic
mechanisms
of
the
pathogenic
process
have
not
been
fully
elucidated,
and
our
progress
is
remarkable
in
the
identification
of
therapeutic
targets
to
achieve
the
control
of
the
inflammatory
process,
yet
our
ability
to
inhibit
the
excessive
bone
formation
is
still
insufficient.
The
study
of
new
bone
formation
in
spondyloarthritis
has
been
mostly
conducted
in
animal
models
of
the
disease
and
only
few
experiments
have
been
done
using
human
biopsies.
The
deregulation
and
overexpression
of
molecules
involved
in
the
osteogenesis
process
have
been
observed
in
bone
cells,
mesenchymal
cells,
and
fibroblasts.
The
signaling
associated
to
the
excessive
bone
formation
is
congruent
with
those
involved
in
the
physiological
processes
of
bone
remodeling.
Bone
morphogenetic
proteins
and
Wnt
pathways
have
been
found
deregulated
in
this
disease;
however,
the
cause
for
uncontrolled
stimulation
remains
unknown.
Mechanical
stress
appears
to
play
an
important
role
in
the
pathological
osteogenesis
process;
nevertheless,
the
association
of
other
important
factors,
such
as
the
presence
of
HLA-B27
and
environmental
factors,
remains
uncertain.
The
present
review
summarizes
the
experimental
findings
that
describe
the
signaling
pathways
involved
in
the
new
bone
formation
process
in
spondyloarthritis
in
animal
models
and
in
human
biopsies.
The
role
of
mechanical
stress
as
the
trigger
of
these
pathways
is
also
reviewed.
©
2015
Société
franc¸
aise
de
rhumatologie.
Published
by
Elsevier
Masson
SAS.
All
rights
reserved.
1.
Introduction
Spondyloarthritis
(SpA)
are
an
interrelated
group
of
rheumatic
diseases
characterized
by
common
clinical
symptoms
and
genetic
similarities.
The
most
important
clinical
features
include
inflammatory
back
pain,
asymmetric
peripheral
oligoarthritis,
pre-
dominantly
in
lower
limbs,
enthesitis,
and
sacroiliitis.
The
SpA
may
also
involve
specific
organs,
such
as
in
anterior
uveitis,
psoriasis
and
chronic
inflammatory
bowel
disease
[1].
Ankylosing
spondyli-
tis
(AS)
is
the
prototype
disease
in
the
study
of
SpA
and
its
hallmark
is
the
new
bone
formation
at
sites
of
enthesitis
[2].
All
SpA
subtypes,
including
AS,
psoriatic
arthritis,
enteropathic
arthritis,
reactive
arthritis,
juvenile
SpA
and
undifferentiated
SpA,
are
characterized
by
arthritis
and
enthesitis
although
their
location,
severity
and
pattern
varies.
Particularly
in
AS,
new
bone
forma-
tion
leading
to
bone
fusion
(ankylosis)
of
sacroiliac
joints
as
well
∗Corresponding
author.
E-mail
address:
dr.cesarpacheco@gmail.com
(C.
Pacheco-Tena).
as
syndesmophytes
that
bridge
the
edges
of
the
vertebral
bodies,
or
enthesophytes
that
proliferate
from
the
edges
of
the
entheses
in
axial
and
appendicular
skeleton.
For
most
SpA
patients,
the
dis-
ease
burden
results
from
the
combination
of
the
bone
inflammation
and
osteoproliferative
structural
changes.
To
date,
the
link
between
inflammation
and
bone
proliferation
processes
remains
elusive
and
a
certain
degree
of
independence
between
both
processes
has
been
proposed
[3,4].
The
new
bone
formation
in
SpA
includes
the
proliferation
of
mesenchymal
precursors,
their
commitment
to
the
bone
lineage
and
eventual
maturation,
and
their
migration
and
eventual
cell
death.
The
osteoproliferation
in
SpA
is
a
complex
process
of
tis-
sue
remodeling
that
shares
similarities
with
joint
remodeling
in
osteoarthritis
[5].
The
distinction
between
physiological
and
patho-
logical
process
of
bone
formation
is
particularly
clear
in
the
SpA:
whereas
the
new
bone
is
formed
in
excess
on
the
outer
surface
of
cortical
bone,
paradoxically,
SpA
patients
develop
osteoporosis
due
to
degradation
of
trabecular
bone
of
the
vertebral
bodies.
These
types
of
divergent
remodeling
in
cortical
and
trabecular
bone
in
SpA
patients
suggest
that
pathological
bone
remodeling
in
SpA
is
clearly
http://dx.doi.org/10.1016/j.jbspin.2015.07.008
1297-319X/©
2015
Société
franc¸
aise
de
rhumatologie.
Published
by
Elsevier
Masson
SAS.
All
rights
reserved.

Please
cite
this
article
in
press
as:
González-Chávez
SA,
et
al.
Molecular
mechanisms
of
bone
formation
in
spondyloarthritis.
Joint
Bone
Spine
(2016),
http://dx.doi.org/10.1016/j.jbspin.2015.07.008
ARTICLE IN PRESS
G Model
BONSOI-4290;
No.
of
Pages
7
2
S.A.
González-Chávez
et
al.
/
Joint
Bone
Spine
xxx
(2016)
xxx–xxx
Table
1
Molecular
mechanisms
of
new
bone
formation
in
SpA
human
samples.
Authors
Sample
Molecular
pathway
Analyzed
molecules
Analysis
technique
Findings
Appel
et
al.,
2010
[5]
Zygapophyseal
joints
of
AS
patients
Bone
differentiation
OPG,
RANKL,
OCN
IHQ
Levels
of
OCN+,
OPG+,
and
RANKL+
osteoblasts
did
not
differ
between
AS
and
OA
Levels
of
OPG+
and
OC+
but
not
RANKL+
osteoblasts
were
significantly
lower
in
controls
compared
to
AS
patients
Osteoblast
activity
is
similar
in
AS
and
OA,
indicating
that
new
bone
formation
is
possibly
a
physiological
function
of
repair
in
both
diseases
Pacheco-Tena
et
al.,
2014
[7]
Synovial
sheaths,
entheses,
and
bone
samples
of
AT
patients
Bone
differentiation
OCN,
OPN,
PTHrP,
BSP,
ALP
IHQ,
IF
OCN,
OPN,
BSP,
and
PTHrP
were
found
in
the
entheseal
and
osteal
tissues
showing
bone
proliferation
OCN
and
OPN
are
over
expressed
in
ankylosis
tarsitis
biopsies
compared
with
normal
controls
OCN
and
OPN
are
expressed
in
a
fibroblast-mesenchymal
phenotype
cells
OCN
and
OPN
in
cells
with
a
fibroblast-mesenchymal
phenotype,
suggesting
the
induction
of
entheseal
cells
toward
an
osteoblast
phenotype
Lories
et
al.,
2005
[8]
Entheseal
biopsies
obtained
from
Achilles
tendons
of
SpA
patients
BMP/Smad
BMP-2,
BMP-6,
BMP-7,
p-Smad
1/5/8,
PCNA
IF
Positive
phosphorylated
smad1/5/8
proliferating
cells
and
negative
chondrocytes
Co-localization
of
phosphorylated
smad1/5/8
and
PCNA
Immunoreactivity
for
BMP2,
BMP7,
and
BMP6
was
recognized
in
proliferating
spindle-shaped
cells
and
in
prehypertrophic
and
mature
chondrocytes,
respectively
Activation
of
BMP
signaling
in
the
proliferating
and
differentiating
cell
population
by
the
presence
of
nuclear
phosphorylated
smad1/5
Wang
et
al.,
2012
[9]
Sacroiliac
joint
tissue
samples
of
AS
patients
TGF-␤1/Smad
TGF-␤1,
p-Smad3,
Smad7,
CTGF,
type
I
and
III
collagen
IHQ
TGF-␤1
and
CTGF
were
over
expressed
in
cytoplasm
of
inflammatory
cells
in
pannus
and
bone
marrow
Meantime,
p-Smad3
was
expressed
in
the
nuclear,
while
Smad7
was
down
expressed
Type
I
and
III
collagen
were
over
expressed
in
bone,
cartilage
and
ligament
tissue
TGF-ˇ1/CTGF
may
play
an
important
role
in
articular
cartilage
fibrosis
and
ossification
of
AS
by
Smad
signal
pathway
Joo
et
al.,
2014
[10]
Blood
BMP
52
genes
related
to
bone
formation
SNPs
Identification
of
new
loci
of
BMP-6
associated
with
radiographic
severity
in
patients
with
AS
Two
SNPs
in
BMP6
(rs270378
and
rs1235192)
were
significantly
associated
with
radiologic
severity
BMP6
is
associated
with
radiographic
severity
in
AS,
supporting
the
role
wingless-type
like/BMP
pathway
on
radiographic
progression
in
AS
Thomas
et
al.,
2013
[11]
Knee
synovial
biopsies
of
AS/SpA
patients
WNT
MMP3,
Dkk-3,
Kremen1
WGEP,
qPCR,
IHQ
Four
hundred
and
sixteen
differentially
expressed
genes
were
identified
that
clearly
delineated
between
AS/SpA
and
control
groups
Pathway
analysis
showed
altered
gene
expression
in
oxidoreductase
activity,
B-cell
associated,
matrix
catabolic,
and
metabolic
pathways
Altered
“myogene”
profiling
was
also
identified
The
inflammatory
mediator,
MMP3,
was
strongly
upregulated
in
AS/SpA
samples
Wnt
pathway
inhibitors,
DKK3
and
Kremen1,
were
downregulated
Supports
the
hypothesis
that
initial
systemic
inflammation
in
SpA
transfers
to
and
persists
in
the
local
joint
environment,
and
might
subsequently
mediate
changes
in
genes
directly
involved
in
the
destructive
tissue
remodelling
ALP:
alkaline
phosphatase;
AS:
ankylosing
spondylitis;
AT:
ankylosing
tarsitis;
BMP:
bone
morphogenetic
protein;
BSP:
bone
sialoprotein;
CTGF:
connective
tissue
growth
factor;
Dkk-3:
dickkopf-related
protein
3;
IF:
immunofluorescence;
IHQ:
immunohistochemistry;
MMP3:
matrix
metalloproteinase
3;
OA:
Osteoarthritis;
OCN:
osteocalcin;
OPG:
osteoprotegerin;
OPN:
osteopontin;
p-:
phosphorylated-;
PCNA:
proliferating
cell
nuclear
antigen;
PTHrP:
parathyroid
hormone-related
protein;
qPCR:
quantitative
polymerase
chain
reaction;
RANKL:
nuclear
factor-kappaB
ligand;
SNPs:
single
nucleotide
polymorphisms;
SpA:
spondyloarthritis;
TGF-␤:
transforming
growth
factor
beta;
WGEP:
whole
genome
expression
profiling.
The
main
findings
are
highlighted
in
italics.
different
from
the
classic
bone
turnover.
Moreover,
the
inflamma-
tion
control
reverts
the
loss
of
trabecular
bone,
but
does
not
seem
to
affect
the
cortical
osteoproliferative
ankylosing
progression
[6].
Studies
of
molecular
mechanisms
of
new
bone
formation
pro-
cesses
in
humans
with
SpA
are
scarce
(Table
1).
Most
of
these
are
limited
to
histological
findings
suggesting
a
contribution
from
the
endochondral
and
membranous
bone
formation
in
the
devel-
opment
of
ankylosis
[5,12–15].
Recently
our
laboratory
reports
the
osteoproliferation
and
abnormal
expression
of
bone
lineage
proteins:
osteocalcin,
osteopontin,
parathyroid
hormone-related
protein
and
bone
sialoprotein
in
biopsies
of
the
entheses
of
ankylos-
ing
tarsitis
patients
compared
with
normal
biopsies.
These
results
suggested
that
ossification
process
may
be,
in
part,
explained
by
the
differentiation
of
mesenchymal
entheseal
cells
toward
the
osteoblastic
lineage
[7].
Molecular
mechanisms
of
new
bone
for-
mation
have
been
studied
mainly
in
animal
models
of
SpA
[16].
The
signaling
pathways
described
in
these
models
are
consistent
with
the
pathways
involved
in
the
normal
process
of
bone
formation,
wherein
the
bone
morphogenetic
proteins
(BMP),
Wnt
and
Hedge-
hog
(Hh)
proteins
have
been
the
most
involved.
Interestingly,
in
the
last
decades,
the
influence
of
mechanical
stress
has
been
linked
to
the
process
of
new
bone
formation
in
SpA,
however,
its
direct
association
with
specific
signaling
pathways
remains
unclear.
2.
Bone
morphogenetic
proteins
signaling
pathways
in
SpA
The
BMPs
are
morphogenic
growth
factors
and
cytokines,
which
were
originally
identified
as
proteins
able
to
induce
the
full
cas-
cade
of
endochondral
bone
formation
[17,18].
Currently
BMPs
are
recognized
as
members
of
the
superfamily
of
transforming
growth
factor
beta
(TGF-␤)
[19].
In
the
essential
process
of
endochondral
ossification,
the
mesenchymal
progenitor
cells
differentiate
into

Please
cite
this
article
in
press
as:
González-Chávez
SA,
et
al.
Molecular
mechanisms
of
bone
formation
in
spondyloarthritis.
Joint
Bone
Spine
(2016),
http://dx.doi.org/10.1016/j.jbspin.2015.07.008
ARTICLE IN PRESS
G Model
BONSOI-4290;
No.
of
Pages
7
S.A.
González-Chávez
et
al.
/
Joint
Bone
Spine
xxx
(2016)
xxx–xxx
3
Fig.
1.
Bone
morphogenetic
proteins
(BMP),
Wnt
and
Hedgehog
(Hh)
signaling
pathways
in
new
bone
formation
in
spondyloarthritides
(SpA).
Wnt
(1–4),
BMP
(5–9)
and
Hh
(10–11)
signaling
pathway
in
new
bone
formation
in
SpA
animal
models.
(1)
Human
tumor
necrosis
factor
(hTNF)
transgenic
mouse
model:
inhibition
of
Dickkopf-related
protein-1
(Dkk-1)
with
anti-TNF
reduces
inflammation,
bone
erosion
and
osteoclast
number
in
the
sacroiliac
joints
without
leading
ankylosis
[21].
(2)
Proteoglycan-induced
arthritis
murine
model
(SpIPG):
Dkk-1
and
sclerostin
(SOST)
expression
levels
are
reduced
in
the
spine
of
SpIPG
mice
compared
to
control
mice
[22].
(3)
hTNF
transgenic
mouse
model:
blocking
of
Dkk-1
with
anti-Dkk-1
has
no
effect
on
sacroilitis,
reduces
bone
erosions
and
osteoclasts
count,
and
promotes
type
X
collagen
expression,
hypertrophic
chondrocytes
formation
and
ankylosis
[23].
(4)
hTNF
transgenic
mouse
model:
treatment
with
R-spondin-1
(Rspo1)
protects
the
damage
associated
with
inflammation
in
bone
and
cartilage
and
preserves
the
structural
integrity
of
joints
[24].
(5)
Model
of
spontaneous
arthritis
in
DBA/1
mouse:
BMP-2,
7
and
6
participate
in
different
stages
of
ankylosing
enthesitis
[8].
(6)
Model
of
spontaneous
arthritis
in
NZB
x
BXSB
mouse
F(1):
BMP-2
is
associated
with
heterotopic
cartilage
and
bone
formation
in
the
enthesis
of
tarsal
bones
[25].
(7)
Model
of
spontaneous
arthritis
in
DBA/1
mouse:
noggin
treatment
inhibits
the
onset
and
progression
of
spontaneous
arthritis
and
modulates
endochondral
bone
formation
in
a
preventive
and
therapeutic
way
[8].
(8)
Mouse
model
DBA/1
-
noggin
(+/LacZ):
endogenous
noggin
reduction
affects
the
progression
of
joint
remodeling
and
slows
the
ossification
process
[26].
(9)
Model
of
spontaneous
arthritis
in
DBA/1
mouse
and
periosteal
cell
culture:
p38
mitogen-activated
protein
kinases
(MAPK)
inhibition
stimulates
ankylosis
(in
vivo)
and
interferes
with
osteochondrogenic
progenitor
cell
differentiation
(in
vitro)
[27].
(10)
Model
of
arthritis
induced
by
serum
transfer
(K/BxN)
in
C57/BL6
mice:
the
blockade
of
the
Hh
pathway
[using
as
target
the
component
of
the
Smoothened
(Smo)
signaling
pathway]
inhibits
the
formation
of
osteophytes;
this
inhibition
does
not
affect
inflammation
but
reduces
bone
destruction
at
local
and
systemic
level
[28].
(11)
Model
of
cross
mb1-Cre
(±)
and
loxP-flanked
Ptc
homologues1
(Ptch1)
mice:
chronic
activation
of
the
Hh
signaling
pathway
in
chondrocytes
of
the
spine
can
trigger
an
ankylosing
morphology
without
contribution
of
immune
cells
[29].
GliA:
activated
glioma-associated
oncogene
family
member;
GliR:
repressor
glioma-associated
oncogene
family
member
Gsk3␤:
glycogen
synthase
kinase-3
␤;
Sufu:
suppressor
of
fused
homologue;
Kif7:
kinesin
family
member
7;
IFT:
intraflagellar
transport.
chondrocytes,
whom
construct
a
cartilage
template
in
which
the
cells
progressively
evolve
towards
their
terminal
state
of
differen-
tiation
(hypertrophy).
Subsequently,
blood
vessels
and
osteoblast
precursors
invade
the
chondrocyte
matrix
and
replace
the
template
by
cartilaginous
bone
[20].
The
best-known
downstream
signals
to
the
primed
BMP
protein
activation
event
are
the
Smad
proteins
and
the
eventual
activa-
tion
of
mitogen-activated
protein
kinases
(MAPKs)
such
as
p38
and
extracellular
signal-regulated
kinases
(ERK)
(Fig.
1).
BMP
signals
are
mediated
by
type
I
and
II
BMP
receptors
and
their
downstream
molecules:
Smad1,
5
and
8.
Phosphorylated
Smad1,
5
and
8
pro-
teins
form
a
complex
with
Smad4.
This
complex
is
translocated
into
the
nucleus
where
it
interacts
with
other
transcription
factors
such
as
runt-related
transcription
factor
2
(Runx2)
in
osteoblasts.
BMP
signaling
is
regulated
at
different
molecular
levels.
(1)
Nog-
gin
and
other
cystine
knots
containing
BMP
antagonists
bind
with
BMP-2,
4
and
7
to
block
BMP
signaling.
(2)
Smad6
binds
type
I
BMP
receptor
and
prevents
Smad1,
5
and
8
activation.
(3)
Tob
inter-
acts
specifically
with
BMP
activated
Smad
proteins
and
inhibits
active
BMP
signaling.
(4)
The
Smad
specific
E3
ubiquitin
pro-
tein
ligase
1
(Smurf1)
interacts
with
Smad1
and
5
and
mediates
the
degradation
of
these
Smad
proteins.
(5)
Smurf1,
also
recog-
nizes
bone-specific
transcription
factor
Runx2
and
mediates
Runx2
degradation.
(6)
Smurf1
also
forms
a
complex
with
Smad6,
which
is
exported
from
the
nucleus
and
targeted
to
type
I
BMP
receptors
for
their
degradation
[19].
Therefore,
the
role
of
Smurf1
as
a
versa-
tile
osteoproliferation
inhibitor
remains
a
matter
of
interest
in
the
perspective
of
diseases
with
abnormal
bone
proliferation
such
as
SpA.
Most
molecular
studies
exploring
the
role
of
the
BMP
in
SpA
come
from
the
spontaneous
model
of
arthritis
in
DBA/1
mice.
These
mice
present
typical
features
of
SpA
peripheral
arthritis
and
enthe-
sitis
limited
to
hind
paws.
The
clinically
evident
arthritis
is
revealed
as
a
rapid
cascade
of
endochondral
ossification
of
the
entheseal
fibrocartilage.
Initially,
in
the
entheses,
mesenchymal
cells
prolifer-
ate
and
differentiate
into
chondrocytes,
and
then
they
hypertrophy
and
subsequently
ossify.
The
process
occurs
in
both
epiphyses
until
complete
fusion
[30].
The
main
findings
in
this
model
are:
•different
BMP
molecules
participate
in
different
stages
of
anky-
losing
enthesitis.
BMP-2
is
present
in
initial
stages
while
BMP-7
and
BMP-6
in
later
stages
[8];
•the
molecule
antagonist
noggin
inhibits
the
onset
and
progres-
sion
of
spontaneous
arthritis
and
endochondral
bone
formation
in
a
preventive
and
therapeutic
way
[8];
•the
endogenous
noggin
affects
the
progression
of
joint
remodel-
ing
and
slows
the
ossification
process
[26];
•p38
MAPK
mediates
the
BMP
signaling
cascades
[27].
Aside
from
the
arthritis
model
in
DBA/1
strain,
the
male
(NZB
×
BXSB)
F(1)
mice
develop
spontaneously
ankylosing
enthe-
sitis/arthritis
in
the
ankle
and
tarsal
joints.
This
ankylosis
is
microscopically
characterized
by
a
marked
proliferation
of
fibroblast-like
cells
positive
for
BMP-2
in
association
with
hetero-
topic
formation
of
cartilages
and
bones
in
hyperplastic
entheseal
tissues
and
subsequent
fusion
of
tarsal
bones
[25].
The
evidence
of
BMP
signaling
in
SpA
human
samples
is
limited.
Lories
et
al.
[8]
analyzed
entheseal
biopsies
from
Achilles
tendons

Please
cite
this
article
in
press
as:
González-Chávez
SA,
et
al.
Molecular
mechanisms
of
bone
formation
in
spondyloarthritis.
Joint
Bone
Spine
(2016),
http://dx.doi.org/10.1016/j.jbspin.2015.07.008
ARTICLE IN PRESS
G Model
BONSOI-4290;
No.
of
Pages
7
4
S.A.
González-Chávez
et
al.
/
Joint
Bone
Spine
xxx
(2016)
xxx–xxx
of
SpA
patients.
BMP-2,
BMP-7,
and
BMP-6
was
detected
in
pro-
liferating
spindle-shaped
cells
and
in
prehypertrophic
and
mature
chondrocytes.
BMP
signaling
activation
was
apparent
in
the
prolif-
erating
and
differentiating
cell
population
revealed
by
the
presence
of
nuclear
phosphorylated
Smad1/5.
Furthermore,
Wang
et
al.
[9]
found
that
TGF-␤1/connective
tissue
growth
factor
(CTGF)
may
play
an
important
role
in
cartilage
ossification
through
Smad
signaling
pathways
in
sacroiliac
joint
tissue
samples.
BMP
also
appears
to
be
a
conferring-risk
gene
to
develop
AS.
Joo
et
al.
[10]
selected
and
genotyped
single
nucleotide
polymorphisms
(SNPs)
for
related
bone
formation
genes.
These
SNPs
were
corre-
lated
with
the
radiographic
severity
of
AS
patients.
The
patients
were
previously
classified
into
two
groups:
•severe
EA,
defined
by
the
presence
of
syndesmophytes
and/or
fusion
in
the
lumbar
or
cervical
spine;
•mild
EA,
defined
by
the
absence
of
any
syndesmophyte
or
fusion.
As
a
result,
a
new
BMP-6
loci
associated
with
radiographic
sever-
ity
was
identified
[10].
3.
WNT
signaling
pathways
in
SpA
Wnt
signaling
pathway
is
essential
for
embryonic
skeletal
development
and
also
plays
a
critical
role
in
homeostasis
and
regen-
eration
of
bones
in
adulthood.
This
pathway
has
been
implicated
in
osteoblastogenesis
and
is
regulated
in
part
by
inflammatory
responses.
Aberrant
regulation
of
Wnt
pathway
has
been
suggested
as
a
key
element
in
the
pathogenesis
of
AS
[31].
The
classification
of
Wnt
signaling
pathways
is
complex.
One
of
the
most
studied
pathways
is
known
as
canonical
signaling
(Fig.
1).
Wnt
stabilizes
the
multifunctional
protein
␤-catenin,
which
is
able
to
activate
transcription
of
several
genes.
The
cytoplasmic
unbound
␤-catenin
is
essential
for
Wnt
signaling
cascades.
In
the
absence
of
Wnt-specific
ligands,
cytoplasmic
␤-catenin
is
kept
at
low
via
constant
targeting
by
a
multiprotein
degradation
com-
plex.
Under
such
conditions,
lymphoid
enhancer-binding
factor/T
cell-specific
(LEF/TCF)
is
associated
with
Groucho
and
represses
target
gene
expression.
On
the
other
hand,
when
Wnt
ligands
bind
to
Frizzled
(FZD)
and
its
co-receptor,
the
low-density
LRP
(lipoprotein
receptor-related
protein)
5/6,
the
receptor
multimer-
izes
and
forms
multiprotein
complexes
called
signalosomes.
In
the
signalosomes,
the
phosphorylation
cascade
that
prevents
␤-
catenin
degradation
takes
place.
Stabilized
␤-catenin
accumulates
in
the
cytoplasm
and
under
specific
stimulus,
translocates
to
the
nucleus
through
a
process
that
in
some
cases
requires
activated
Ras
mediation.
Nuclear
␤-catenin
displaces
Groucho
and
forms
a
complex
with
the
B-cell
lymphoma-9
protein
(BCL9),
Pygopus,
histone
modifier
CBP
(CREB
[cAMP
response
element-binding]-
binding
protein)
and
tissue
specific
transcriptional
activators.
This
whole
complex
converts
LEF/TCF
from
a
transcriptional
repressor
to
an
activator,
which
activates
gene
expression
[32].
In
addition
to
intracellular
␤-catenin
regulation,
an
extracellular
regulation
of
Wnt
is
known.
Family
members
of
FZD
related
proteins,
Cer-
berus
and
WIF,
directly
bind
and
antagonize
the
extracellular
Wnt.
Moreover,
members
of
Dickkopf
(Dkk)
family
and
sclerostin
(SOST)
bind
to
LRP
co-receptor
and
antagonize
canonical
Wnt
signaling
(Fig.
1).
Higher
total
levels
of
Dkk-1
(Wnt
antagonist)
are
detectable
in
the
serum
of
AS
patients
if
compared
with
levels
of
rheumatoid
arthritis
patients
or
healthy
individuals.
Nevertheless,
the
activity
of
serum
Dkk-1
is
reduced
in
the
patients
with
AS,
since
it
fails
to
inhibit
␤-catenin
activation
in
Jurkat
T-cells
induced
by
AS
patients’
serum.
Moreover,
the
addition
of
anti-Dkk-1
does
not
increase
Wnt
signaling
if
serum
from
AS
patients
is
used
in
counterpart
to
that
from
healthy
control
serums
[33].
In
our
opinion,
extrapolating
functional
results
from
Wnt-␤-catenin
signaling
chain
from
T-cells
to
mesenchymal-osteoblastic
cells
should
be
done
with
reserve.
In
similar
terms
Diarra
et
al.
[21]
have
shown
that
the
serum
levels
of
total
Dkk-1
(measured
by
ELISA)
may
differ
to
those
of
active
Dkk1
if
the
activity
is
measured
by
the
binding
of
Dkk-1
to
chimeric
LRP-6
coated
plates
assay.
If
only
the
functional
levels
of
Dkk-1
are
con-
sidered,
the
serum
concentration
of
Dkk-1
might
indeed
be
lower
in
patients
with
AS
if
compared
to
RA
or
healthy
control
serums.
Furthermore,
it
has
been
determined
that
in
AS
patients
with
no
new
syndesmophyte
formation,
the
serum
levels
of
functional
Dkk-
1
are
higher
than
those
with
new
growth
syndesmophytes.
The
Dkk-1
levels
correlated
with
the
SOST
levels,
suggesting
that
block-
ade
of
Wnt
signaling
suppresses
new
bone
formation
and
likewise
syndesmophyte
growth
and
ankylosis
[34].
This
may
be
a
physiological
antagonistic
mechanism.
Moreover,
it
has
been
found
that
expression
of
Wnt
pathway
inhibitors,
Dkk-
3
and
Kremen1,
are
downregulated
in
knee
synovial
biopsies
of
AS
patients
in
comparison
with
rheumatoid
arthritis
and
normal
biopsies
[11].
The
relationship
between
Wnt
signaling
and
the
bone
forma-
tion
in
SpA
have
been
studied
in
more
detail
in
animal
models,
particularly
in
the
transgenic
mouse
of
human
tumor
necrosis
fac-
tor
(hTNF).
This
mouse
has
been
studied
as
a
model
of
synovial
inflammation
and
progressive
joint
destruction
but
without
new
bone
formation.
The
TNF-␣
is
a
Dkk-1
inducer
and
the
effect
of
both
(TNF
and
Dkk)
on
Wnt
as
key
regulator
of
bone
remodeling
has
been
studied.
The
inhibition
of
Dkk-1
with
anti-TNF
antibody
effec-
tively
reduces
inflammation,
bone
erosion,
and
osteoclast
numbers
in
the
sacroiliac
joints
preventing
ankylosis.
Moreover,
the
spe-
cific
blockade
of
Dkk-1
with
anti-Dkk-1
antibodies
has
no
effect
on
the
inflammatory
changes
in
the
sacroiliac
but
does
reduce
bone
erosions
and
interestingly
promotes
expression
of
type
X
collagen,
formation
of
hypertrophic
chondrocytes
and
ankylosis
of
sacroiliac
joints
[21,23].
In
addition
to
the
role
of
Dkk
regulators
in
Wnt
signaling,
R-spondin
family
(Rspo)
has
shown
the
ability
to
amplify
the
Wnt/␤-catenin
activity.
Experiments
in
TNF-␣
transgenic
mice
showed
that
Rspo1
prevents
the
osteocartilaginous
damage
induced
by
inflammation
and
preserves
the
joint’s
structural
integrity
[24].
Excessive
bone
formation
associated
with
the
reduction
of
Wnt
inhibitors
has
also
been
proven
in
the
proteoglycan-induced
spondylitis
(SpIPG)
mouse
model.
Dkk-1
and
SOST
expression
levels
are
decreased
in
the
spine
of
SpIPG
mice
compared
to
control
mice
[22].
4.
Hedgehog
signaling
pathways
in
SpA
The
Hh
signaling
pathway
plays
many
important
roles
in
bone
development
and
homeostasis.
This
pathway
requires
a
distinct
cell
organelle,
the
cilium.
The
Hh-binding
protein,
Ptc
homologues1
(Ptch1),
is
located
in
the
cilium,
whereas
transmembrane
protein
Smoothened
(Smo)
is
kept
outside
of
the
cilium
in
the
absence
of
Hh
ligands.
Glioma-associated
oncogene
family
member
(Gli)
is
phos-
phorylated
by
kinases,
such
as
protein
kinase
A
(PKA),
casein
kinase
1
(CK1)
and
glycogen
synthase
kinase-3
␤
(Gsk3␤),
which
promote
the
processing
of
the
repressor
form
(GliR)
in
a
␤-Trcp-dependent
manner.
Hh
signaling
is
blocked.
When
Hh
ligands
bind
to
Ptch1,
Smo
inhibition
is
relieved.
Ptch1
exits
from
the
cilium,
whereas
Smo
is
translocated
to
cilium.
The
repressor
form
of
the
Gli
(GliR),
suppressor
of
fused
homologue
(Sufu),
and
kinesin
family
mem-
ber
7
(Kif7)
complex
travel
from
the
base
of
the
cilium
to
the
top
via
intraflagellar
transport
(IFT).
Kif7
blocks
the
function
of
Sufu
at
the
top
of
the
cilium.
Gli
is
not
processed
and
is
maintained
its
active
form
(GliA).
Activated
Gli
travels
from
the
top
of
the
cilium
to

Please
cite
this
article
in
press
as:
González-Chávez
SA,
et
al.
Molecular
mechanisms
of
bone
formation
in
spondyloarthritis.
Joint
Bone
Spine
(2016),
http://dx.doi.org/10.1016/j.jbspin.2015.07.008
ARTICLE IN PRESS
G Model
BONSOI-4290;
No.
of
Pages
7
S.A.
González-Chávez
et
al.
/
Joint
Bone
Spine
xxx
(2016)
xxx–xxx
5
Fig.
2.
Bone
mechanosensing
and
remodeling
processes.
Bone
remodeling

is
strictly
regulated
by
communication
between
bone
cells:
osteoclasts,
osteoblasts
and
osteocytes.
During
bone
remodeling,
bone
resorption
by
osteoclasts
precedes
bone
formation
by
osteoblasts
.
Osteocytes
are
able
to
sense
their
mechanical
environments
through
different
mechanosensors
including
cytoskeleton,
focal
adhesions
and
primary
cilia
.
Different
bone
cells
(osteocytes,
osteoblasts)
and
enthesis
cells
(mesenchymal
stem
cells,
fibroblasts)
can
overexpress
molecules
of
differentiation
leads
to
a
bone
lineage
in
response
to
mechanical
stress
(MS)
.
the
cytoplasm
via
IFT
and
translocates
to
the
nucleus
to
transcript
target
genes
thereby
activating
Hh
signaling
[35].
Hh
relationship
in
pathologic
osteogenesis
is
not
conclusive
in
humans;
however
some
murine
models
of
arthritis
suggest
poten-
tial
role
in
this
process.
In
the
model
of
arthritis
induced
by
serum
transfer
(K/BxN)
in
C57/BL6
mice,
the
blockade
of
the
Hh
path-
way
(using
as
target
the
component
of
the
Smo
signaling
pathway)
inhibits
the
formation
of
osteophytes;
this
inhibition
does
not
affect
inflammation
but
reduces
bone
destruction
at
local
and
systemic
level
[28].
More
recently,
Dittmann
et
al.
[29]
blocked
pathways
inhibition
for
Hh
in
chondrocytes
of
mice
obtained
from
crossing
the
strain
mb1-Cre
(±)
and
loxP-flanked
Ptch1.
These
findings
indicate
that
chronic
activation
of
the
Hh
signaling
pathway
in
chondrocytes
of
the
spine
can
trigger
an
ankylosing
morphology
without
contribu-
tion
of
immune
cells.
Therefore,
the
authors
suggest
that
cartilage
destruction
and
loss
of
integrity
of
the
axial
union
can
result
from
defects
of
the
chondrocyte
intrinsic
defects.
5.
Mechanical
load
and
new
bone
formation
in
SpA
The
bone
is
constantly
renewed
by
the
balanced
action
of
forma-
tion
and
bone
resorption,
which
takes
place
mainly
on
the
surface.
This
process
known
as
“bone
remodeling”
is
important
not
only
for
the
maintenance
of
normal
bone
mass
and
strength,
but
also
for
mineral
homeostasis.
Bone
remodeling
is
strictly
regulated
by
communication
between
bone
cells:
osteoclasts,
osteoblasts
and
osteocytes.
During
bone
remodeling,
bone
resorption
by
osteoclasts
precedes
bone
formation
by
osteoblasts
(Fig.
2)
[36].
The
osteo-
cytes
are
the
most
abundant
and
long-lived
cell
in
the
bone
tissue.
Through
cell
synapses,
osteocytes
maintain
contact
with
each
other
and
with
other
types
of
cells
on
the
surface
of
the
bone
matrix.
These
contacts
form
a
dynamic
and
active
network
of
cells
that
regulate
bone
homeostasis.
The
integrated
network
of
osteocytes
is
essential
for
maintaining
normal
function
of
bone
tissue
[37].
The
bone
and
particularly
the
osteocytes
have
been
identified
as
highly
mechanosensitive
structures.
Mechanotransduction
is
the
process
by
which
the
physical
mechanical
stimuli
are
trans-
lated
to
biochemical
responses.
This
process
is
vital
to
maintaining
bone
integrity
in
physiological
conditions.
Mechanical
translated
signals
activate
and
suppress
multiple
signaling
cascades
that
regulate
bone
formation
and
resorption.
Cells
are
able
to
sense
their
mechanical
environments
through
different
mechanosensors
including
cytoskeleton,
focal
adhesions
and
primary
cilia
(Fig.
2).
The
cytoskeleton
provides
a
structural
framework
for
the
cell,
wherein
myosin
and
actin
are
organized
into
contractile
struc-
tures
that
generate
tension.
This
framework
allows
the
mechanical
stimuli
perception
that
induces
osteogenesis.
Some
surface
pro-
teins
link
the
cytoskeleton
binding
proteins,
including
several
integrins,
which
anchor
the
osteocyte’s
cell
membrane
to
the
extra-
cellular
matrix
and
form
focal
adhesions.
These
focal
adhesions
are
associated
with
a
variety
of
signaling
molecules.
The
forces
trans-
mitted
from
the
bone
matrix
to
the
focal
adhesions
are
important
for
mechanical
load
(ML)-induced
osteogenesis.
The
primary
cil-
ium
is
unique
and
motionless
like
an
antenna
extending
from
the
cell
to
the
extracellular
space
[38].
The
signaling
mechanisms
are
complex
since
the
variety
of
mechanical
signals
that
can
maintain
or
disrupt
cellular
homeostasis
through
transcriptional
regulation
of
growth
factors,
matrix
proteins
and
inflammatory
mediators
in
both
normal
and
pathological
conditions
[39].
There
is
abundant
evidence
that
ML
mediates
the
activation
of
signaling
pathways
that
result
in
cell
differentiation
and
bone
formation
(Table
2).
Therefore,
it
is
possible
to
hypothesize
that
these
tissue
mechanosensitive
mechanisms
may
be
relevant
in
the
pathological
process
of
new
bone
formation
in
SpA.
In
fact,
many
of
the
findings
in
the
molecular
study
of
mechanosensitive
signaling
involved
BMP
[43,45,46,49,51,54,55]
and
Wnt
[40,41,50],
which
have
been
the
most
described
in
the
pathological
process
of
SpA.
Mechanotransduction
and
BMP
signaling
are
strongly
intercon-
nected
through
an
elaborate
network
of
active
signaling
during
bone
development
and
homeostasis
[56].
Mechanical
compres-
sion
stress
induces
chondrogenic
differentiation
via
BMP
signaling
[49].
Additionally,
ML
quickly
represses
TGFß
activity
in
osteocytes,
resulting
in
reduced
Smad2
and
Smad3
activity.
Mechanosensitive
TGFß
signaling
regulation
is
essential
to
mechanical
stress-induced
bone.
Indeed,
loss
of
TGFß
sensitivity
compromises
bone
anabolic
response
to
ML
[46].
Another
associated
BMP
signaling
is
p38
MAPK,
however,
it
has
been
reported
that
p38
MAPK
is
not
involved
in
osteogenic
differentiation
during
early
response
to
ML
[51].
Osteocytes
may
also
control
the
osteoblastic
differentiation
of
mesenchymal
precursors
in
response
to
ML
via
Wnt
signaling
[57].
Rolfe
et
al.
[41]
identified
mechanosensitive
genes
during
skeletal
development
in
a
mouse
model
whose
mechanical
stimulation
was
altered.
Wnt
signaling
showed
the
greatest
alteration,
and
it
was
found
that
several
cytoskeletal
components
mediate
this
response.
ML-induced
bone
formation
via
Wnt
has
also
been
demonstrated
in
deficient
of
SOST
(Sost
−/−)
transgenic
mice.
SOST
deficiency
can
increase
bone
formation
when
ML
is
increased
[40].
In
addition
to
ML-induced
osteogenesis
on
bone
cells
[42–46],
it
has
been
demonstrated
that
ML
induces
bone
differentiation

Please
cite
this
article
in
press
as:
González-Chávez
SA,
et
al.
Molecular
mechanisms
of
bone
formation
in
spondyloarthritis.
Joint
Bone
Spine
(2016),
http://dx.doi.org/10.1016/j.jbspin.2015.07.008
ARTICLE IN PRESS
G Model
BONSOI-4290;
No.
of
Pages
7
6
S.A.
González-Chávez
et
al.
/
Joint
Bone
Spine
xxx
(2016)
xxx–xxx
Table
2
Bone
formation
induced
by
mechanical
loading.
Study
model
Stimuli
Effect
Signaling
mediators
Reference
SOST
deficient
transgenic
mice
(Sost−/−)
Mechanical
load
Bone
formation
SOST
independent
pathway
(MAR
and
BFR/BS
periosteum)
[40]
Mutant
mice
model
impaired
of
mechanical
stimulation
by
absence
of
limb
skeletal
muscle
Poor
mechanical
stimulation
Cytoskeletal
rearrangement
and
cell
signaling
Wnt4
[41]
Primary
calvarial
cells
derived
from
knockout
Er␤
mice
(BERKO)
Mechanical
load
Mechanical
signaling
regulation
of
osteoblasts
Estrogen
receptor
␤
[42]
Mouse
osteoblast
(MC3T3-E1) Cyclic
stretching Increased
expression
of
osteogenic
genes
(collagen
type
I
and
osteopontin)
ERK
[43]
Osteoblasts
Oscillatory
stress
Osteoblast
stimulation
TRPM7
[44]
Osteoblast
(MC3T3-E1)
Fluid
shear
stress
Osteogenic
differentiation
BMP2,
ALP,
RunX2,
SP7,
type
I
collagen,
integrin
␤1
[45]
Osteocytes
Mechanical
load Bone
formation SOST,
TGF␤[46]
Mesenchymal
stem
cells
derived
from
rat
periosteum
(P-MSCs)
Negative
pressure
Stem
cell
proliferation
and
osteogenic
differentiation
Integrin
␤5
[47]
Mesenchymal
stem
cells
derived
from
human
bone
marrow
Magnetic
force
Increased
expression
of
osteogenic
genes
PDGFR␣,
Integrin
␣v␤3
[48]
Mesenchymal
stem
cells
derived
from
rat
bone
marrow
Compression
stress
Chondrogenic
differentiation
BMP
[49]
Mesenchymal
stem
cells
(C3H10T1/2)
Oscillatory
fluid
flow
Osteogenic
differentiation
Wnt5a,
Ror2,
N-cadherin,
␤-catenin
[50]
Mesenchymal
stem
cells
Continuous
mechanical
strain
Early
osteogenesis
and
increased
expression
of
osteogenic
genes
ALP,
type
I
collagen,
OPN,
p38
MAPK
[51]
Human
derived
fibroblast Mechanical
stretch Osteogenic
differentiation OCN,
ALP,
type
I
collagen
[52]
Human
derived
fibroblast
Cyclic
stretch
Increased
expression
of
osteogenic
genes
OCN,
ALP,
type
I
collagen
[53]
Adipose-derived
stem
cells
Uniaxial
cyclic
tensile
strain
Osteogenic
differentiation
BMP-2,
Runx2,
OCN
[54]
Adipose-derived
stem
cells
Cyclic
mechanical
stretch
Adipogenesis
inhibition
and
osteogenesis
stimulation
PPAR-␥,
Runx2,
Pref-1,
ERK1/2
[55]
ALP:
alkaline
phosphatase;
BFR:
bone
formation
rates;
BMP:
bone
morphogenetic
protein;
BS:
bone
surfaces;
ERK:
extracellular
signal-regulated
kinases;
MAPK:
mitogen-
activated
protein
kinases;
MAR:
mineral
apposition
rate;
OCN:
osteocalcin;
OPN:
osteopontin;
PDGFR␣:
platelet-derived
growth
factor
receptor
␣;
PPAR-␥:
peroxisome
proliferator-activated
receptor
gamma;
Pref-1:
preadipocyte
factor
1;
RunX2:
runt-related
transcription
factor
2;
SOST:
sclerostin;
SP7:
Sp7
transcription
factor
7;
TGF␤:
transforming
growth
factor
beta;
TRPM7:
transient
receptor
potential
cation
channel,
subfamily
M,
member
7.
in
progenitor
cells
and
other
cell
lines.
Mechanical
stimulation
induces
bone
differentiation
in
mesenchymal
stromal
cells
via
inte-
grins
[47,48]
BMP
[49,51]
and
Wnt
[50]
signaling.
Furthermore,
ML
induce
the
expression
of
bone
lineage
proteins
in
fibroblast
[52,53].
In
adipocytes,
ML
induces
osteogenesis
while
it
simultaneously
inhibits
adipogenesis
[54,55].
Studies
of
ML-induced
osteogenesis
in
SpA
are
very
limited.
The
enthesitis
based
model
for
SpA
pathogenesis,
suggests
the
involvement
of
biomechanical
stress
in
the
pathogenic
process
of
inflammation
in
the
entheses
[58].
Moreover,
tasks
involving
flex-
ion,
torsion,
stretching
and
exposure
to
whole
body
vibration
are
the
professional
activities
associated
with
greater
functional
limi-
tations
and
greater
radiographic
damage
in
SpA
patients
[59].
Although
experimental
evidence
that
supports
ML-induced
osteogenesis
in
SpA
is
poor,
Jacques
et
al.
[60]
recently
found
that
new
bone
formation
in
SpA
is
influenced
by
ML.
The
effect
of
ML
on
bone
formation
was
evaluated
using
the
collagen
antibody-
induced
arthritis
(CAIA)
model.
This
model,
induced
in
susceptible
DBA/1
mice,
exhibited
enthesitis
at
7
days
of
immunization,
which
eventually
became
a
destructive
polyarthritis.
Osteophyte
forma-
tion
occurred
after
resolution
of
the
inflammatory
phase.
When
clinical
arthritis
was
established,
a
group
of
mice
were
tail
sus-
pended
to
avoid
ML
in
hind
paws,
while
another
group
remained
in
standard
cages.
Interestingly,
tail
suspended
mice
with
minimum
clinical
arthritis
did
not
develop
osteophytes
and
they
presented
significantly
smaller
osteophytes
when
they
had
greater
arthritis
development.
New
bone
formation
occurred
distant
from
the
joint
surface
specifically
in
the
entheseal
sites.
This
is
the
first
specific
study
that
highlights
the
ML
role
in
entheseal
new
bone
formation
in
an
SpA
model.
In
conclusion,
our
understanding
of
the
mechanisms
underlying
the
excessive
bone
formation
in
SpA
is
incomplete.
Likely,
this
bone
formation
is
triggered
by
the
inflammatory
response,
but
in
later
disease
stages,
some
degree
of
independence
between
both
could
exist.
Most
of
the
studies
dealing
with
the
link
between
inflamma-
tion
and
osteoproliferative
responses
are
based
on
rodent
models
of
arthritis-enthesitis.
On
those,
several
signaling
pathways
including
BMP’s,
WNT
and
Hg
have
demonstrated
its
potential
to
reproduce
SpA
key
features
of
bone
formation
and
entheseal
abnormalities.
In
humans,
there
is
scarce
evidence
that
implicates
the
BMP
and
WNT
pathway’s
involvement
in
SpA.
As
mentioned
by
Benjamin
and
others,
mechanical
demand
explains
the
pattern
of
enthesitis
in
SpA
altogether
with
the
presence
of
fibrocartilage.
Several
pathways
bridge
mechanopro-
pioception
to
osteogenesis;
indeed,
a
fundamental
stimulus
in
bone
restoration
and
housekeeping
is
dependent
on
mechanical
demand.
Inflammation,
however,
in
a
physiological
perspective
plays
no
role
in
osteogenesis.
Our
limited
understanding
of
the
potential
roles
of
HLA-B27
in
regard
to
mechanopropioception
complicates
our
understanding
of
the
potential
roles
of
this
HLA
antigen;
however,
its
interaction
with
the
involved
mediators
in
the
intrinsic
bone
biology
are
worthy
of
being
explored.
The
relative
independence
of
the
inflammatory
and
the
osteo-
proliferative
processes
provides
an
opportunity
to
improve
our
therapeutical
frontiers.
It
is
now
assumed
that
anti-TNF
therapy
succeeds
at
a
larger
extent
preventing
ankylosis
in
AS
patients
with
non-chronic
vertebral
lesions.
Therefore,
in
patients
with
established
disease,
anti-TNF
may
not
suffice
to
prevent
dis-
ease
progression
despite
controlling
inflammation
and
improving
symptoms.
Hopefully,
our
understanding
of
the
altered
bone
biol-
ogy
in
SpA
could
empower
us
to
completely
abolish
disease
progression.

Please
cite
this
article
in
press
as:
González-Chávez
SA,
et
al.
Molecular
mechanisms
of
bone
formation
in
spondyloarthritis.
Joint
Bone
Spine
(2016),
http://dx.doi.org/10.1016/j.jbspin.2015.07.008
ARTICLE IN PRESS
G Model
BONSOI-4290;
No.
of
Pages
7
S.A.
González-Chávez
et
al.
/
Joint
Bone
Spine
xxx
(2016)
xxx–xxx
7
Disclosure
of
interest
The
authors
declare
that
they
have
no
competing
interest.
References
[1]
Braun
J,
Sieper
J.
Spondyloarthritides.
Z
Für
Rheumatol
2010;69:425–32.
[2]
Golder
V,
Schachna
L.
Ankylosing
spondylitis:
an
update.
Aust
Fam
Physician
2013;42:780–4.
[3]
Lories
RJU,
Dougados
M.
Inflammation
and
ankylosis:
still
an
enigmatic
rela-
tionship
in
spondyloarthritis.
Ann
Rheum
Dis
2012;71:317–8.
[4]
Maksymowych
WP,
Elewaut
D,
Schett
G.
Motion
for
debate:
the
development
of
ankylosis
in
ankylosing
spondylitis
is
largely
dependent
on
inflammation.
Arthritis
Rheum
2012;64:1713–9.
[5]
Appel
H,
Maier
R,
Loddenkemper
C,
et
al.
Immunohistochemical
anal-
ysis
of
osteoblasts
in
zygapophyseal
joints
of
patients
with
ankylosing
spondylitis
reveal
repair
mechanisms
similar
to
osteoarthritis.
J
Rheumatol
2010;37:823–8.
[6]
Lories
RJU,
Schett
G.
Pathophysiology
of
new
bone
formation
and
ankylosis
in
spondyloarthritis.
Rheum
Dis
Clin
N
Am
2012;38:555–67.
[7]
Pacheco-Tena
C,
Pérez-Tamayo
R,
Pineda
C,
et
al.
Bone
lineage
proteins
in
the
entheses
of
the
midfoot
in
patients
with
spondyloarthritis.
J
Rheumatol
2015;42:630–7.
[8]
Lories
RJU,
Derese
I,
Luyten
FP.
Modulation
of
bone
morphogenetic
protein
signaling
inhibits
the
onset
and
progression
of
ankylosing
enthesitis.
J
Clin
Invest
2005;115:1571–9.
[9]
Wang
Q,
Zeng
P,
Cai
Y,
et
al.
Expression
of
transforming
growth
factor
beta
1
and
connection
tissue
growth
factor
in
ankylosing
spondylitis.
Beijing
Da
Xue
Xue
Bao
2012;44:244–9.
[10]
Joo
YB,
Bang
S-Y,
Kim
T-H,
et
al.
Bone
morphogenetic
protein
6
polymorphisms
are
associated
with
radiographic
progression
in
ankylosing
spondylitis.
PloS
One
2014;9:e104966.
[11]
Thomas
GP,
Duan
R,
Pettit
AR,
et
al.
Expression
profiling
in
spondyloarthropa-
thy
synovial
biopsies
highlights
changes
in
expression
of
inflammatory
genes
in
conjunction
with
tissue
remodelling
genes.
BMC
Musculoskelet
Disord
2013;14:354.
[12]
Appel
H,
Kuhne
M,
Spiekermann
S,
et
al.
Immunohistologic
analysis
of
zygapophyseal
joints
in
patients
with
ankylosing
spondylitis.
Arthritis
Rheum
2006;54:2845–51.
[13]
Appel
H,
Kuhne
M,
Spiekermann
S,
et
al.
Immunohistochemical
analy-
sis
of
hip
arthritis
in
ankylosing
spondylitis:
evaluation
of
the
bone-
cartilage
interface
and
subchondral
bone
marrow.
Arthritis
Rheum
2006;54:
1805–13.
[14]
Appel
H,
Loddenkemper
C,
Grozdanovic
Z,
et
al.
Correlation
of
histopatholog-
ical
findings
and
magnetic
resonance
imaging
in
the
spine
of
patients
with
ankylosing
spondylitis.
Arthritis
Res
Ther
2006;8:R143.
[15]
Franc¸
ois
RJ,
Gardner
DL,
Degrave
EJ,
et
al.
Histopathologic
evidence
that
sacroiliitis
in
ankylosing
spondylitis
is
not
merely
enthesitis.
Arthritis
Rheum
2000;43:2011–24.
[16]
Braem
K,
Lories
RJ.
Insights
into
the
pathophysiology
of
ankylosing
spondylitis:
Contributions
from
animal
models.
Joint
Bone
Spine
2012;79:243–8.
[17]
Urist
MR.
Bone:
formation
by
autoinduction.
1965.
Clin
Orthop
2002;395:4–10.
[18]
Wozney
JM,
Rosen
V,
Celeste
AJ,
et
al.
Novel
regulators
of
bone
formation:
molecular
clones
and
activities.
Science
1988;242:1528–34.
[19]
Chen
D,
Zhao
M,
Mundy
GR.
Bone
morphogenetic
proteins.
Growth
Factors
Chur
Switz
2004;22:233–41.
[20]
Carter
S,
Braem
K,
Lories
RJ.
The
role
of
bone
morphogenetic
proteins
in
anky-
losing
spondylitis.
Ther
Adv
Musculoskelet
Dis
2012;4:293–9.
[21]
Diarra
D,
Stolina
M,
Polzer
K,
et
al.
Dickkopf-1
is
a
master
regulator
of
joint
remodeling.
Nat
Med
2007;13:156–63.
[22]
Haynes
KR,
Pettit
AR,
Duan
R,
et
al.
Excessive
bone
formation
in
a
mouse
model
of
ankylosing
spondylitis
is
associated
with
decreases
in
Wnt
pathway
inhibitors.
Arthritis
Res
Ther
2012;14:R253.
[23]
Uderhardt
S,
Diarra
D,
Katzenbeisser
J,
et
al.
Blockade
of
Dickkopf
(DKK)-1
induces
fusion
of
sacroiliac
joints.
Ann
Rheum
Dis
2010;69:592–7.
[24]
Krönke
G,
Uderhardt
S,
Kim
K-A,
et
al.
R-spondin
1
protects
against
inflamma-
tory
bone
damage
during
murine
arthritis
by
modulating
the
Wnt
pathway.
Arthritis
Rheum
2010;62:2303–12.
[25]
Abe
Y,
Ohtsuji
M,
Ohtsuji
N,
et
al.
Ankylosing
enthesitis
associated
with
upre-
gulated
IFN-gamma
and
IL-17
production
in
(BXSB
×
NZB)
F(1)
male
mice:
a
new
mouse
model.
Mod
Rheumatol
Jpn
Rheum
Assoc
2009;19:316–22.
[26]
Lories
RJU,
Daans
M,
Derese
I,
et
al.
Noggin
haploinsufficiency
differentially
affects
tissue
responses
in
destructive
and
remodeling
arthritis.
Arthritis
Rheum
2006;54:1736–46.
[27]
Braem
K,
Luyten
FP,
Lories
RJU.
Blocking
p38
signalling
inhibits
chondrogenesis
in
vitro
but
not
ankylosis
in
a
model
of
ankylosing
spondylitis
in
vivo.
Ann
Rheum
Dis
2012;71:722–8.
[28]
Ruiz-Heiland
G,
Horn
A,
Zerr
P,
et
al.
Blockade
of
the
hedgehog
pathway
inhibits
osteophyte
formation
in
arthritis.
Ann
Rheum
Dis
2012;71:400–7.
[29]
Dittmann
K,
Wuelling
M,
Uhmann
A,
et
al.
Inactivation
of
patched1
in
murine
chondrocytes
causes
spinal
fusion
without
inflammation.
Arthritis
Rheumatol
Hoboken
NJ
2014;66:831–40.
[30]
Lories
RJU,
Matthys
P,
de
Vlam
K,
et
al.
Ankylosing
enthesitis,
dactylitis,
and
ony-
choperiostitis
in
male
DBA/1
mice:
a
model
of
psoriatic
arthritis.
Ann
Rheum
Dis
2004;63:595–8.
[31]
Corr
M.
Wnt
signaling
in
ankylosing
spondylitis.
Clin
Rheumatol
2014;33:759–62.
[32]
Yu
J,
Virshup
DM.
Updating
the
Wnt
pathways.
Biosci
Rep
2014;34:e00142.
[33]
Daoussis
D,
Liossis
S-NC,
Solomou
EE,
et
al.
Evidence
that
Dkk-1
is
dysfunctional
in
ankylosing
spondylitis.
Arthritis
Rheum
2010;62:150–8.
[34]
Heiland
GR,
Appel
H,
Poddubnyy
D,
et
al.
High
level
of
functional
dickkopf-1
predicts
protection
from
syndesmophyte
formation
in
patients
with
ankylosing
spondylitis.
Ann
Rheum
Dis
2012;71:572–4.
[35]
Yang
J,
Andre
P,
Ye
L,
et
al.
The
Hedgehog
signalling
pathway
in
bone
formation.
Int
J
Oral
Sci
2015;7:73–9.
[36]
Nakashima
T.
Stress
and
cell
communication
between
bone
cells.
Clin
Calcium
2013;23:1595–603.
[37]
Zeng
H,
Bai
D,
Han
X-L.
The
biological
function
of
osteocytes.
Sheng
Li
Ke
Xue
Jin
Zhan
2014;45:100–4.
[38]
Chen
JC,
Jacobs
CR.
Mechanically
induced
osteogenic
lineage
commitment
of
stem
cells.
Stem
Cell
Res
Ther
2013;4:107.
[39]
Knapik
DM,
Perera
P,
Nam
J,
et
al.
Mechanosignaling
in
bone
health,
trauma
and
inflammation.
Antioxid
Redox
Signal
2014;20:970–85.
[40]
Morse
A,
McDonald
M,
Kelly
N,
et
al.
Mechanical
load
increases
in
bone
forma-
tion
via
a
sclerostin-independent
pathway.
J
Bone
Miner
Res
2014;29:2456–67.
[41]
Rolfe
RA,
Nowlan
NC,
Kenny
EM,
et
al.
Identification
of
mechanosensitive
genes
during
skeletal
development:
alteration
of
genes
associated
with
cytoskeletal
rearrangement
and
cell
signalling
pathways.
BMC
Genomics
2014;15:48.
[42]
Castillo
AB,
Triplett
JW,
Pavalko
FM,
et
al.
Estrogen
receptor-␤
regulates
mechanical
signaling
in
primary
osteoblasts.
Am
J
Physiol
Endocrinol
Metab
2014;306:E937–44.
[43]
Matsui
H,
Fukuno
N,
Kanda
Y,
et
al.
The
expression
of
Fn14
via
mechanical
stress-activated
JNK
contributes
to
apoptosis
induction
in
osteoblasts.
J
Biol
Chem
2014;289:6438–50.
[44]
Roy
B,
Das
T,
Mishra
D,
et
al.
Oscillatory
shear
stress-induced
calcium
flickers
in
osteoblast
cells.
Integr
Biol
Quant
Biosci
Nano
Macro
2014;6:289–99.
[45]
Mai
Z,
Peng
Z,
Wu
S,
et
al.
Single
bout
short
duration
fluid
shear
stress
induces
osteogenic
differentiation
of
MC3T3-E1
cells
via
integrin
␤1
and
BMP2
signaling
cross-talk.
PloS
One
2013;8:e61600.
[46]
Nguyen
J,
Tang
SY,
Nguyen
D,
et
al.
Load
regulates
bone
formation
and
Sclerostin
expression
through
a
TGF␤-dependent
mechanism.
PloS
One
2013;8:e53813.
[47]
Zhu
J,
Yu
A,
Qi
B,
et
al.
Effects
of
negative
pressure
wound
therapy
on
mesenchy-
mal
stem
cells
proliferation
and
osteogenic
differentiation
in
a
fibrin
matrix.
PloS
One
2014;9:e107339.
[48]
Hu
B,
El
Haj
AJ,
Dobson
J.
Receptor-targeted,
magneto-mechanical
stimulation
of
osteogenic
differentiation
of
human
bone
marrow-derived
mesenchymal
stem
cells.
Int
J
Mol
Sci
2013;14:19276–93.
[49]
Wang
Y,
Wang
J,
Bai
D,
et
al.
Cell
proliferation
is
promoted
by
compressive
stress
during
early
stage
of
chondrogenic
differentiation
of
rat
BMSCs.
J
Cell
Physiol
2013;228:1935–42.
[50]
Arnsdorf
EJ,
Tummala
P,
Jacobs
CR.
Non-canonical
Wnt
signaling
and
N-
cadherin
related
beta-catenin
signaling
play
a
role
in
mechanically
induced
osteogenic
cell
fate.
PloS
One
2009;4:e5388.
[51]
Zhang
P,
Wu
Y,
Dai
Q,
et
al.
p38
MAPK
signaling
pathway
is
not
involved
in
osteogenic
differentiation
during
early
response
of
mesenchymal
stem
cells
to
continuous
mechanical
strain.
Mol
Cell
Biochem
2013;378:19–28.
[52]
Zhang
W,
Wei
P,
Chen
Y,
et
al.
Downregulated
expression
of
vimentin
induced
by
mechanical
stress
in
fibroblasts
derived
from
patients
with
ossification
of
the
posterior
longitudinal
ligament.
Eur
Spine
J
2014;23:2410–5.
[53]
Xu
P,
Ye
X,
Xu
G,
et
al.
Morphologic
and
apoptotic
changes
of
cervical
posterior
longitudinal
ligament
fibroblasts
in
response
to
in
vitro
mechanical
stretch.
Zhonghua
Yi
Xue
Za
Zhi
2011;91:786–9.
[54]
Yang
X,
Gong
P,
Lin
Y,
et
al.
Cyclic
tensile
stretch
modulates
osteogenic
differ-
entiation
of
adipose-derived
stem
cells
via
the
BMP-2
pathway.
Arch
Med
Sci
AMS
2010;6:152–9.
[55]
Yang
X,
Cai
X,
Wang
J,
et
al.
Mechanical
stretch
inhibits
adipogenesis
and
sti-
mulates
osteogenesis
of
adipose
stem
cells.
Cell
Prolif
2012;45:158–66.
[56]
Kopf
J,
Paarmann
P,
Hiepen
C,
et
al.
BMP
growth
factor
signaling
in
a
biome-
chanical
context.
Biofactors
2014;40:171–87.
[57]
Galli
C,
Passeri
G,
Macaluso
GM.
Osteocytes
and
WNT:
the
mechanical
control
of
bone
formation.
J
Dent
Res
2010;89:331–43.
[58]
McGonagle
D,
Stockwin
L,
Isaacs
J,
et
al.
An
enthesitis
based
model
for
the
pathogenesis
of
spondyloarthropathy.
Additive
effects
of
microbial
adjuvant
and
biomechanical
factors
at
disease
sites.
J
Rheumatol
2001;28:2155–9.
[59]
Ward
MM,
Reveille
JD,
Learch
TJ,
et
al.
Occupational
physical
activities
and
long-term
functional
and
radiographic
outcomes
in
patients
with
ankylosing
spondylitis.
Arthritis
Rheum
2008;59:822–32.
[60]
Jacques
P,
Lambrecht
S,
Verheugen
E,
et
al.
Proof
of
concept:
enthesitis
and
new
bone
formation
in
spondyloarthritis
are
driven
by
mechanical
strain
and
stromal
cells.
Ann
Rheum
Dis
2014;73:437–45.